# Discrepancy between Chat and Responses API Image Generation tool output **URL:** \ **Category:** API\ **Tags:** chatgpt, image-generation, gpt-5, gpt-image-1\ **Created:** [October 29, 2025, 6:58pm UTC](https://community.openai.com/t/discrepancy-between-chat-and-responses-api-image-generation-tool-output/1364399 "2025-10-29T18:58:59Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![graciey](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/graciey/32/513834_2.png) [@graciey](https://community.openai.com/u/graciey)\ **Post date:** [October 29, 2025, 6:58pm UTC](https://community.openai.com/t/discrepancy-between-chat-and-responses-api-image-generation-tool-output/1364399/1 "2025-10-29T18:58:59Z")
Hi everyone, I’m running into a big difference in image generation quality between **Chat** and the **OpenAI Responses API** (gpt-5) with **ImageGeneration tool** (gpt-image-1), even when using identical prompts and images. **Goal:** I’m building a workflow to generate packshot-quality images of clothing items and accessories. The input is usually a photo of a garment (either taken by a user in their closet or sourced from a website where the item is shown on a model). The desired output is a clean, e-commerce–style product photo — like what you’d see on professional retail sites. **Setup:** - **Model** : gpt-5 via the **Responses API** - **Tool config:** - Same prompt and same input image each time - Input image passed via `content[]` as We’re explicitly using the **Responses API** with the **Image Generation** tool to replicate the behavior of Chat as closely as possible. **Issue** When I run the prompt _in Chat_, the results are beautiful — sharp lighting, realistic textures, clean backgrounds, and accurate product details. But when I run the _exact same prompt and image_ through the API, the quality drops a lot: - Details about the garment are wrong - Lighting and shadowing are inconsistent - Image looks less professional I need to process **thousands of images** , so I can’t rely on Chat manually — I really need API-level consistency that matches Chat’s quality. Interestingly, **my colleague** , who’s been generating a _larger volume_ of images via Chat (with the same model and prompt), consistently gets _better-quality outputs_ than I do. This makes me think it has something to do with personalization. **Questions:** - Has anyone else noticed this difference between Chat and the Responses API results? - Are there hidden differences in how the Chat interface calls the image tool vs how the API would call the image tool (e.g., preprocessing, better/personalized automatic prompt expansion, system context)? Below is an example (you can see the Chat result looks great, the API result added a zipper and got the collar shape wrong). Thanks for any insights! ![chat_api_issues](https://us1.discourse-cdn.com/openai1/original/4X/f/a/4/fa44be635b61315b3c9512d08414ba8bf8ad836e.png) ---
**Author:** ![simoerba](https://avatars.discourse-cdn.com/v4/letter/s/b5e925/32.png) [@simoerba](https://community.openai.com/u/simoerba)\ **Post date:** [January 20, 2026, 9:58am UTC](https://community.openai.com/t/discrepancy-between-chat-and-responses-api-image-generation-tool-output/1364399/2 "2026-01-20T09:58:49Z")
Hello! I got the same problem with my specific task: generating realistic images of insects on a flowerbed. The problem you highlighted is still present with `gpt-image-1.5`: using the API I get mid-quality images of the insects I want to generate, but using the UI the flower quality, depth of field, lighting, etc… are better. I have to generate a bunch of images, so relying on the UI is completely unfeasible. Have you solved this problem? ---
**Author:** ![mikeb360](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/mikeb360/32/673798_2.png) [@mikeb360](https://community.openai.com/u/mikeb360)\ **Post date:** [February 27, 2026, 2:40am UTC](https://community.openai.com/t/discrepancy-between-chat-and-responses-api-image-generation-tool-output/1364399/3 "2026-02-27T02:40:37Z")
I too am having this issue. I get better results when using a chat dialog, but through the api my results are hugely inconsistent. I’ve been trying to solve it all day and not really any further than when I began :-/. I get a lot of random 500 server errors as well, with no changes to the prompt. Both issues are driving me nuts.